Skip to content

Latest commit

 

History

History
110 lines (85 loc) · 4.58 KB

README.md

File metadata and controls

110 lines (85 loc) · 4.58 KB
Raw

UMatter Bot

UMatter - A bot for instant appreciation via Mattermost

Project Summary:

This project aims to make mattermost more rewarding by appreciating peers in Mattermost channels right when and where great work is happening.

Hackathon Proposal:

This is a submission to Mattermost Hackathon.

Hackathon Proposal can be found here - Proposal

Note: One change in the proposal - Negative points are not implemented. After some thinking, awarding negative points is a bad way of communication. Hence, this feature was not implemented.

Features:

(Click on the feature for more info)

  • Give Appreciation - Quickly give appreciation to the user right there in the channel itself.
  • Create a Company Value - Create a company value as an emoji and tag to the appreciation posts
  • Top Peers - Users can see top performing teammates in their respective channels wrt to a date range
  • Self Statistics - See the statistics about yourself
  • Channel Statistics - See the statistics in a public/private channel

Development:

Setting up the environment

This is the manual way of setting up the environment and the app

  • Programming Language: Python 3.7

  • Database: MySQL 8.0

  • Create a database by name - "umatter"

  • Run the script in db/init.sql

  • Mattermost Version: v5.20

  • Virtual Environment

    • Create an anaconda environment.
    • Activate the conda environment.
    • Run the command pip install requirements.txt

  • Create a Bot Account in Mattermost with system admin access. Copy the Token and assign it to the corresponding config class MM_BOT_TOKEN. Follow the image below.

  • Similarly if you are running the app in local, register a slash command with the url http://127.0.0.1:5000/v1/index/. Follow the image below.

  • Configuration Variables

    • Flask Settings
    Variable Name Possible Values Purpose
    APP_NAME u-matter rest service Name of the application
    APP_PORT 5000 port where the flask app runs
    APP_HOST localhost host url for the flask app
    DEBUG True/False To print the debug logs
    • Mysql Settings
    Variable Name Possible Values Purpose
    MYSQL_HOST localhost MySql host url
    MYSQL_USER root MySql user name
    MYSQL_PASSWORD root MySql User Password
    MYSQL_DB umatter MySql Database to Use
    MYSQL_PORT 3306 MySql Port
    • Mattermost settings
    Variable Name Possible Values Purpose
    DAILY_POINT_LIMIT 10 Max points given by an user per day
    PER_TRANSACTION_POINT_LIMIT 5 Max points given by an user for every appreciation
    WEEKLY_THRESHOLD 40 Weekly threshold for reward. Anyone crossing this value for 7 days are considered for reward
    MM_SCHEME http/https To use SSL encryption for MM connectivity or not
    MM_URL localhost Mattermost host
    MM_PORT 8065 Mattermost port
    MM_BOT_TOKEN wegffdasdfgrearfdgb Token once you create a bot account on MM
    MM_SLASH_TOKEN sefgrrwefgrq3wreg Token once you register a slash command in Mattermost
    • Log File Path
      • LOG_FILE_PATH = "event_logs.log"

  • All of the above configuration variables are in the config.py. You need to set the values accordingly.

  • default configuration is set to production. In order to change it, set the environment variable FLASK_CONFIG to development or testing

  • In Order to run the application manually, python run.py

  • If everything works fine, you should see the logs printing as Serving on http://0.0.0.0:5000

Docker

This is the dockerized way of setting up the environment and the app.

  • Here, set the mattermost url, port, bot token, slash command token.
  • Run the command docker-compose up
  • If everything works fine, you should be able to access the slash command in your mattermost UI.

Troubleshooting Common Problems:

  1. Not able to access mattermost API - Set the Bot Token correctly
  2. Not able to connect to database - Check if the mysql settings in the config are correct

Enhancements Needed:

  • Automate the creation of the bot account
  • Increase the statistics of channel
  • Better format messages

Pending Tasks:

  1. Unit testing
  2. behavioral testing
  3. Integration testing
  4. Coverage

A big thanks to Christian Vaelor for his mattermost driver.